基于改进的SSA优化SVR的某工业园区短期负荷预测

Short-Term Power Load Forecasting for an Industrial Park Based onImproved SSA and Optimized SVR

为实现不规律、波动性大、不确定性的电力负荷数据高精度预测,提出了一种使用小波包分解(WPD)与麻雀搜索算法(SSA)来优化支持向量回归(SVR)的短期负荷预测方案。该方案使用WPD将原始负荷序列分解成多个各异的小波动分量,将分解后的各组数据分别输入SSA优化后的SVM模型,并将得到的多个各异的小波动分量分别经模型预测出的结果进行相加得到最后取得的预测结果。结果表明:该方案能较好拟合整个测试集上的实际预测点位,适合于电力系统短期负荷的准确预测,证实了该模型的有效性和优越性。

To achieve high-precision prediction of irregular,highly volatile,and uncertain power load data,a short-term load forecasting scheme with using wavelet packet decomposition(WPD)and sparrow search algorithm(SSA)is proposed to optimize support vector regression(SVR).Firstly,WPD is used to decompose the original load into multiple distinct small fluctuation components.Then,each group of decomposed data is inputted into the SSA optimized SVM model.Finally,the obtained multiple distinct small fluctuation components are added up to the predicted results of the model to obtain the final prediction result.The results show that this scheme can well fit the actual predicted points on the entire test set,and is suitable for accurate short-term load prediction of the power system,confirming the effectiveness and superiority of the model.